bug in the nano package?

[volumetricsteve]

I just tried to install nano from the GUI and it acted like nothing happened....confused and looking for an error message, I went back into command line mode and used tce to manually download and install the nano.tcz file, which appeared to work.

When I went to run nano, I got an error that libncurses.so.5 couldn't be found, so it seemed that there was some dependency that didn't auto-load when nano was pulled down, the same way the tc-install.tcz package pulled down a ton of other things automatically.

Going back into tce, I looked up the "tree" for nano, and found mention of "ncurses.tcz" and "ncurses-common.tcz" so I downloaded and installed those manually, now nano works fine.   Out of the bunch of tiny core apps/extensions I've downloaded thus far, this is the only one that didn't auto-download it's dependencies.  Is that to be expected or did I do something wrong?

tl;dr  I got it working now, it was just a weird work-around process compared to other installs

[gerald_clark]

tce-load -w nano works for me. The dep file contains ncurses.tcz.

[volumetricsteve]

weird...I bet I did my install weird/wrong somehow, once I got the gui working, I noticed the terminal icon wasn't showing up in the wbar, so I'm guessing other things were out of place as well.

[gerald_clark]

There is no default gui terminal.
Did you install aterm?

[volumetricsteve]

Oh I just installed LXTerminal, but good to know.

[PDP-8]

Same here with TC 7.2 in 2017 ..

Hint: it might have to do with running cli ncurses based extensions in X loaded installed with "ondemand".  I'm still investigating.

Example:
Generic install of 7.2 on a i586 Pentium (Dell Inspiron 8100).  Frugal install to hard disk.  ext2 fs

Loaded a few extensions: htop, cfdisk, and nano as "ondemand".

When called up from the gui Aterm terminal, they fail or the process runs away.  I quicky use xkill or kill to stop them.

At first I thought just using a cli-based program in a cli environment (ctrl-alt-f1) would do the trick, but they sometimes just ran away there too.

Long story short, I did on-demand extension maintenance, and removed them from the ondemand list.  Now they seem to be stable if I'm in the cli, or in an aterm.

The irony is watching htop run away with 100% cpu.

It's probably me, so I'm going to do more research about optimizing ondemand.

[Misalf]

AFAICT, The On-Demand feature was meant to run GUI apps.
Not all CLI apps work as expected when run through On-Demand scripts (i.e. when ran the first time).
I found that the  tce-run  script provides better/expected results with CLI apps, compared to the  launchApp()  function. I believe it's related to something about reloading the environment. Starting a new shell usually (always?) makes things work as expected.
Maybe it would make sense to give it another look. Perhaps even ditching  launchApp()  and just use  tce-run ?
Further investigation required..

[PDP-8]

Ah, thanks for that - will give it a try.

I think that when I evaluate my own environment, unless I am ram-starved, there is no need for me to use ondemand - a few extra k here and there won't hurt for programs I use often.

Heh, there too - how many times do I need to run cfdisk on a daily basis?   A standard load would be fine anyway.

Or, just use a simpler tool like fdisk instead.  Use the built-ins in the unix toolkit approach instead of relying on a captive user interface yadda yadda.

TC is making me snatch-the-pebble from Ken Thompson's hand.

[wksilva]

Te file  nano.tcz.dep is not OK

The file nano.tcz.dep  says "ncurses.tcz" but nano wants "ncurses5.tcz"

 $ ldd $(which nano)
   libncursesw.so.5 => not found

To solve this wile the package is not fixed:

 $ tce-load -iw ncurses5.tcz

[Juanito]

It looks OK from here:

http://tinycorelinux.net/8.x/x86/tcz/nano.tcz.dep

[wksilva]

But not for x86_64:

    http://mirror.cedia.org.ec/tinycorelinux/8.x/x86_64/tcz/nano.tcz.dep

[Juanito]

If you don't state the tc version or the architecture, it will be assumed that you refer to the latest version for x86

updated nano posted